A mixed-reality display presents computer-generated virtual objects that can “augment” the view of a real-world environment surrounding a user. For instance, a user can place a three-dimensional virtual item, such as a cup, to appear as if the virtual item is positioned on a real-world, “physical” object, such as a table.
Although some existing computing devices allow users to position virtual objects on or near physical objects in a real-world environment, features that enable users to interact with such virtual objects can be limiting. For instance, in the above-described example involving the virtual cup that is positioned to appear on the real-world table, not all systems can manage the relationship between the cup and the table when the table is moved. When a particular device, such as a head-mounted device, is used to track the movement of the table, a number of errors can occur when the table is moved out of the view of the head-mounted device. In this scenario, the virtual object can become disassociated from the real-world object, and a computer may lose its ability to display the virtual object with the right orientation or at the right position. Such a result can completely undermine the purpose of a mixed-reality environment, which is to position items in a precise manner to augment a real-world view. Such issues are exacerbated when the mixed-reality environment is displayed during a communication session, such as a live meeting or a communication broadcast to many users.